Automatic Dialog Flow Generation for Virtual Assistants
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for generating transactional flows for virtual assistants are inefficient and prone to human errors, requiring significant manual effort and resources, and lack a consistent methodology for scaling and improving flow performance.
Innovation Solution
A method and system for automatically generating transactional dialog flows for virtual assistants by analyzing a tree structure and selecting candidate nodes based on similarity scores, using processors to compute and aggregate similarity sub-scores for node attributes, and generating new flows that are descendants of selected parent nodes, reducing the need for manual intervention and improving flow performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual methods are used to generate transactional flows for virtual assistants, then the flows can be created with human oversight and understanding, but the process requires significant manual effort and resources, is time-consuming, and is prone to human errors
Solution Approach 1:
The system copies patterns and structures from existing predefined transactional flows to generate new flows automatically. By analyzing the tree structure of dialog nodes and identifying similar patterns, the system replicates successful flow structures without manual recreation, significantly reducing time while maintaining quality through pattern-based replication
Solution Approach 2:
The system performs self-service by automatically generating transactional flows without requiring manual intervention for each new flow. The automated flow generation system analyzes the virtual assistant's tree structure, identifies candidate completion nodes, computes similarity scores, and generates flows independently, eliminating the need for continuous manual flow creation
2Ease of manufacture
If manual methods are used to generate transactional flows, then human understanding can guide the process, but the process requires significant manual effort and resources
Solution Approach 1:
The system automates the flow generation process to serve itself, eliminating the need for manual effort in creating each new transactional flow. The automated system analyzes the tree structure, computes similarity scores, and generates flows independently, dramatically improving productivity while maintaining ease of use through automated processes
Solution Approach 2:
The system copies existing flow patterns and structures to create new flows automatically, making the creation process as easy as selecting and adapting proven patterns. This copying mechanism maintains ease of manufacture by using established templates while dramatically improving productivity through automated replication and adaptation
3Adaptability or versatility
If traditional methods are used for generating transactional flows, then consistency can be maintained through human judgment, but there is lack of consistent methodology for scaling and improving flow performance
Solution Approach 1:
The system segments the flow generation process into distinct automated components: analyzing the tree structure of dialog nodes, identifying candidate completion nodes, computing similarity scores between candidates and existing flows, selecting optimal nodes based on scores, and generating the final flow. This segmentation enables scaling while managing complexity through modular automated processes
Solution Approach 2:
The system uses copying of existing flow patterns as a scalable methodology. By computing similarity scores and replicating proven flow structures, the system provides a consistent, scalable approach that can be applied repeatedly without increasing complexity, as the same automated processes handle each new flow generation
4Productivity
If automated flow generation is implemented, then time and resources are reduced, but the system requires complex algorithms for computing similarity scores and selecting nodes
Solution Approach 1:
The complex processing system is segmented into manageable automated components: tree structure analysis, candidate node identification, similarity score computation, node selection based on scores, and flow generation. Each segment handles a specific task with defined algorithms, making the overall complex system manageable while maintaining high productivity through automated execution of all segments
Data Source
AI summary
Disclosed herein are methods and systems for generating automatically transactional dialog flows for a virtual assistant based on a set of predefined (labeled) transactional flows defining user interaction paths to accomplish one or more tasks. The virtual assistant is represented by a tree structure comprising a plurality of dialog nodes where each transactional flow comprises a respective subset of the nodes descending from a parent node to one or more completion nodes. New transactional flows may be generated automatically based on similarity with the predefined transactional flows, by selecting a candidate node and one of the predefined transactional flows comprising a completion node most similar to the candidate completion node and further selecting a candidate parent node most similar to the parent node of the selected predefined transactional flow. The new transactional flows may be then generated to comprise all descendant nodes of the candidate parent node.